What didn’t work. Part 1: coding

What I’m listening to while I type: Hunky Dory

I’ve had two failed start-ups and maybe I’ll have more over the next decade, so I have a vested interest in understanding what makes it more likely that one web-tech start-up will succeed and another one fail.

I’m hoping the research I’ve begun will come up with answers born out of the sweat and tears of entrepreneurs, investors and helpers who have been there and worn the start-up t-shirt. However, I’m going to get the ball rolling by talking about the things that my own experience has taught me.

Starting a web-tech business is like starting a band. You might play some gigs and get a decent following, but the likelihood is that within a couple of years you’ll have split up, given up, or reformed as a tribute band.

But… a tiny few of you will start selling records CDs downloads, and a smaller number will get a contract, and an even smaller number will get some Radio 1 airplay, and a tiny, tiny number of you will become Adele.

It’s like dreaming of winning the lottery, but winning because you’re actually good at something, not because the tooth fairy said your £1 would be the one. It’s not (just) about the money, it’s about standing in the sunbeam of peer recognition.

I digress. Where was I?

Ok, these are the five things that, based on my experience, I believe make web-tech start-up success more likely. There’ll be different and hopefully better lists as I talk to some of you and, if you’ve got suggestions, do feel free to comment or get in touch.

My list below might see obvious, but obvious doesn’t mean do-able. If there was always a direct link between us knowing what we needed to do and actually doing it, we’d all be a lot more successful and the planet wouldn’t be dying around us. So, my list:

  1. Do your own coding (at least to start with)
  2. Don’t only spend your own money
  3. Twitter chatter isn’t enough, good press and advertising lasts longer
  4. It’s the business model, not the business plan, that matters most
  5. Don’t get distracted.

I’ll cover the five over this and the next four posts. Here’s number one.

1. Do your own coding (at least to start with).

I can’t, and I really believe that’s the main reason my last start-up sank.

Expecting programmers to build something in code that you’ve drafted in Word, Excel, or as a notepad scribble – not good.

Not understanding programming languages well enough to be able to tell a) whether that’s a great programmer or great bullshitter sitting in front of you, or b) whether they’re standing at a coding crossroad and might be about to chose the road to obsolescence  –  also not good.

Not understanding the job of coding well enough to know how long you can realistically expect a programmer to take to do that “tweak” you thought of last night as you were falling asleep. Not good.

I’m not the only one who thinks being able to code matters for web-tech entrepreneurs. Here’s Fred Wilson,  of Union Square Ventures, saying the same thing earlier this year:

“Be technical, get technical, or find people you can be business partners with who are. Even if you are not going to do these things by yourself, it’s helpful to learn. It’s better to learn it to talk to technical people, and to evaluate if somebody technical is good or not.”

Stephan Schmidt, of blog codemonkeyism and his own failed start-up, puts “write code” as his number one advice to CTOs of start-ups. While serial entrepreneur David Cummings, blogs that every CEO should learn to code because it will make them “a better leader”. For reasons that include:

‘”You understand the technical architecture and trade-offs of different product decisions. You can call B.S. if a technical person pushes back on something being too difficult/time consuming.”

Being able to code, even a little, means being able to sketch your idea in a way that programmers will be able to follow and run with. It means being able to cobble together a working demo of your idea and be talking to possible investors, while every one else is still writing the brief and interviewing developers.

It means being the pack leader of your technical terriers.

Cesar MilanPlus coding twists the brain, in a useful way. Twitter co-founder Jack Dorsey (talking to Lisa Chow) gave a programmer’s perspective on why they had to come up with the name Twitter before they even started to code the concept:

‘CHOW: At the point that you decided on the name, how much of the idea of the company had been already discussed and decided?
DORSEY: I really wanted a name before we started work on the code because if you don’t name the code the same thing, then it becomes extremely confusing and then you have go in and rename all these variables and all these functions, and it’s just a mess. So I wanted the name before we even started work.’

And in case you wondered, Evan can code too. And obviously Mark, and Larry and Sergey and Bill and Dennis and … (although the other Mark can’t so I do need to interview him).

So what do you do if you’ve got the ideas but not the code skills? I’d suggest looking these options:

Share this post...Share on FacebookShare on Google+Tweet about this on TwitterShare on LinkedInShare on RedditEmail this to someone

Leave a Reply

Your email address will not be published. Required fields are marked *